A new low-loss diode, a self-biased channel diode, is described. In order to achieve a two-terminal operation by using a self-bias, the shunt electrode of the source and gate of the conventional DMOSFET with a floating body is adopted in this diode. By utilizing a DMOSFET, this proposed diode attains a high breakdown voltage compared with a lateral MOSFET. In this device, forward conduction is caused by the self-gate bias created by applying a positive voltage to the shunt electrode of the source and gate with respect to the drain. The direction of forward conduction is opposite to that of forward conduction in a conventional DMOSFET. In the reverse-bias state, the reverse current is very small without any bias owing to the shunt of the source and the gate electrode. In this report, the operational mechanism and electrical characteristics of the device fabricated for the proposed diode are discussed. From the experimental results, it is clear that at room temperature, the on-state voltage of the proposed diode is between that of the Ti-SBD and Cr-SBD. The simulated I-V characteristics are consistent with the measured values. From the simulation results, the proposed diode, in which a thin gate oxide layer and a high integration density of the DMOSFET cell are used, shows lower power loss in the temperature range 25-75°C than does the Cr-SBD. At high temperatures the power loss in the proposed diode is lower than that observed in the Ti-SBD and Cr-SBD which easily fall into thermal runaway.
Today, various diagnosis methods exist to diagnose bearing failure in industries such as artificial neural network (ANN), fuzzy logic, genetic algorithm and statistical analysis. This paper proposes the development of ANN model of induction motor bearing (IMB) fault diagnosis using different network types and structures. In this case Feedforward Neural Network (FFNN) and Elman Network types with Multiple Input Single Output (MISO) and Multiple Input Multiple Output (MIMO) structures were studied. The raw data used in this work was obtained from the Case Western Reserve University website in form of vibration signal. During pre-processing stage, Fast Fourier Transform (FFT) and enveloping techniques were applied to raw data before it was fed to features extraction stage. A total of 16 features were extracted from both time and frequency domain respectively. Subsequently, a distance evaluation technique was used for features selection, where 9 salient features were selected for ANN fault diagnosis. In the development of ANN fault diagnosis, FFNN and Elman Network were utilized with training algorithm of Levenberg Marquart Backpropagation. The result indicates the performance of classified IMB fault by using MIMO Elman structure which was better compared to other combination structures.
This paper proposes a new robust load position servo system without mechanical vibrations and offset angles for use in industrial machines. Generally, industrial machines use sensors mounted on the actuator side because it is difficult to connect signal lines if the sensor is mounted on the load side. Therefore, the load position response does not reach the desired position owing to the effect of an external disturbance force, which is called offset angle. The proposed robust servo system uses the estimated load information as the effect of the external disturbance force is suppressed. The proposed position servo system includes a state disturbance observer and a robust speed control system and uses a coprime factorization controller. The control performance and effectiveness of the proposed robust control system based on the estimated information are confirmed by the experimental results.
This paper presents a performance evaluation and a simple speed control method of an asymmetrical parameter type two-phase induction motor drive using a three-leg VSI (Voltage Source Inverter). The two-phase induction motor is adapted from an existing single-phase induction motor resulting in impedance unbalance between main and auxiliary windings. The unbalanced two-phase inverter outputs with orthogonal displacement based on a SPWM (Sinusoidal Pulse Width Modulation) method are controlled with appropriate amplitudes for improving the motor performance. Dynamic simulation of the proposed drive system is given. A simple speed controller based on a slip regulation method is designed. The overall system is implemented on a DSP (Digital Signal Processor) board. The validity of the proposed system is verified by simulation and experimental results.
This paper proposes a new high-frequency current correlation method with a high-frequency integral-type PLL as a versatile rotor-phase estimation method using stator current caused by high-frequency voltage injection for sensorless drive of salient-pole permanent-magnet synchronous motors. The proposed method has the following advantages. 1) It can be applied for the voltage injection under various conditions. 2) In principle, it can be used for the accurate estimation of rotor phase over a wide speed range. 3) The high-frequency integral-type PLL is extended to higher order for versatility. 4) An analytical method for designing the PLL has been established. 5) It guarantees stable phase estimation. 6) It ensures that the high-frequency noises do not appear in the estimated rotor phase.
This paper proposes a correlation-based model-free controller design method for the tuning of linear time invariant (LTI) multivariable controllers. In the proposed method, the controller parameters are tuned directly by using sets of input/output data obtained independently from the controlled plant; the number of sets is equal to the number of the input dimension of the controlled plant so that the closed-loop output approximates the output of the given reference model. In addition, if the controllers are parameterized linearly with respect to the controller parameters, optimal control parameters can be obtained by the least-squares method. The effectiveness of the proposed method is experimentally confirmed by using a two-input two-output system, namely, a tension-and-speed control apparatus.
Generally, under a wet railway track condition, the axle adhesion coefficient of a train gradually increases from the head of a rolling stock. The improvement of adhesion characteristics is important in an electric motor coach. In order to suppress the slip/skid phenomenon, we have already proposed an anti-slip/skid re-adhesion control system based on a disturbance observer and sensor-less vector control. Moreover, we have confirmed that this system drives the train with the high adhesion force utilization ratio. This paper discusses the theoretical characteristic of the axle adhesion coefficient distribution of a rolling stock, which is based on the axle adhesion coefficient data of Shinkansen. In order to maintain the desired driving force performance, this paper proposes a new planning torque distribution control based on the axle adhesion coefficient distribution. This paper points out that the proposed method well maintains the desired driving force performance by the numerical simulation results.
This paper focuses on the stability issue of power hardware-in-the-loop (PHIL) simulation of inductor coupled systems. Main factors which influence the stability of this type system are analyzed and verified through experiments. A method to resolve the instability problem of the PHIL simulation without decreasing the simulation accuracy is also proposed. Moreover, a PHIL simulation of a gas engine cogeneration system with a matrix converter is introduced and the application of proposed method to improve stability is demonstrated.
This paper proposes a noncontact spinning mechanism that spins a levitated object (in this case, an iron ball) using four rotary permanent disk magnets. When the object was stably levitated in the vertical direction by a magnetic suspension system, the disk magnets spun the suspended object in the horizontal direction by means of the remanent magnetization on the surface of the suspended object. In this mechanism, the disk magnets are arranged around the levitated object, and the magnetic poles are arranged in a parallel configuration. In this paper, the noncontact spinning principle of the noncontact spinning mechanism is discussed. Second, a prototype and a control system of the spinning system are introduced. Third, the rotational torque of the iron ball is calculated using a mathematical model and IEM (integral element method) analysis. Finally, the spinning experiments of the iron ball's velocity step response to the disk magnets and the relationship between the input and output velocities are examined, and the results are shown and discussed. All results indicate that a suspended object can be spun using this noncontact spinning mechanism.
This report proposes a new processing method for automatically detecting the states of road surfaces from tire noises of passing vehicles. In addition to multiple indicators of the signal features in the frequency domain, we propose a few feature indicators in the time domain to successfully classify the road states into four categories: snowy, slushy, wet, and dry states. The method is based on artificial neural networks. The proposed classification is carried out in multiple neural networks using learning vector quantization. The outcomes of the networks are then integrated by the voting decision-making scheme. Experimental results obtained from recorded signals for ten days in the snowy season demonstrated that an accuracy of approximately 90% can be attained for predicting road surface states using only tire noise data.
This paper discusses heat distribution on a workpiece in a zone-control induction heating (ZCIH) system. The system consists of two or more split working coils and multi inverter units, and adjust the coil currents to control the heat or temperature distribution on the workpiece. This paper theoretically reveals the relation between the coil currents and the heat distribution, and proposes a simple method to calculate the heat distribution using a three-dimensional resistance matrix. The validity of the developed theory is confirmed by performing both numerical computation based on the finite element method (FEM) and experiments using a 1.5-kW six-zone ZCIH laboratory setup.
There is a possibility of the realization of high output power density converters by introducing flying capacitor multilevel topologies. However, as the number of the levels increases, the number of circuit components will exceed the practical limit of the implementation. Thus, from the practical viewpoint, the main circuit and related gate drive circuits including floating gate power supplies should be integrated. In this letter, charge pump circuits, already proposed for conventional 2-level converters, are extended to the floating gate power supplies for the flying capacitor multilevel converters.
This paper deals with applying the constant dc voltage control based algorithm to the current balancer in single-phase three-wire secondary distribution systems. The power flows into a three-leg inverter, which performs the current balancer, are discussed, and then we show that only the reactive power flows into the three-leg inverter. This demonstrates that the required-capacity of the dc capacitor can be reduced as compared to that of the active power quality compensator used in the electrified railways. This reduced-capacity dc capacitor results in the rapid and stable balancing. The basic principle of the proposed method is discussed in detail, and then confirmed by digital computer simulation. Digital computer simulation results demonstrate that the balanced source currents with unity power factor are obtained remaining the unbalanced load currents conditions.